snopes.com  

Go Back   snopes.com > SLC Central > Soapbox Derby

Reply
 
Thread Tools Display Modes
  #1  
Old 31 May 2014, 04:16 PM
Steve Steve is offline
 
Join Date: 19 October 2002
Location: Charleston, SC
Posts: 4,847
Default Economists-An astonishing record – of complete failure

In the 2001 issue of the International Journal of Forecasting, an economist from the International Monetary Fund, Prakash Loungani, published a survey of the accuracy of economic forecasts throughout the 1990s. He reached two conclusions. The first was that forecasts are all much the same. There was little to choose between those produced by the IMF and the World Bank, and those from private sector forecasters. The second conclusion was that the predictive record of economists was terrible. Loungani wrote: “The record of failure to predict recessions is virtually unblemished.”

http://www.ft.com/intl/cms/s/2/14e32...#axzz33JHtHM4n
Reply With Quote
  #2  
Old 31 May 2014, 05:56 PM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

Sure, we're bad at seeing recessions in advance. I'd say that's at least as much due to bad zombie economics ideas being in vogue over the pat 30 years as anything else, but there is a deeper, worse idea at the core: a lot of freshwater economists argue that there is no such thing as recessions. That isn't to say that economies never contract 2 quarters in a row, it's that when this happens it's as close to random as anything can be, because if it wasn't random we'd have been better able to predict them by now.

I'm sorry, but this is just dumb, kind of like saying that there are no hurricanes because we can't predict when and where they'll occur 6 months in advance. In a sense, it's even worse than that, because weather forecasting is a whole lot better studied (and a whole whole lot less politicized) than macroeconomics. There is a class of economists who insist that they have developed the ideal, Platonic model for how markets work and any deviation from this is a failure of reality, not a failure of the model. This is poisonous thinking that's worse than merely being wrong.

What I think these folks need to do is understand that even though we can't predict recessions now, that doesn't mean that they don't occur, and that once we accept that they do in fact occur we can seek to study how to mitigate their effects. I think a lot of saltwater economists - Paul Krugman and Brad DeLong, for instance - have already begun to ask these questions and as a result their responses - spending in the face of recession, for example, and cutting back when the economy is performing well - flies in the face of what the freshwater crop that's by and large currently in power is prescribing (witness Europe's "austerity measures" which many of the saltwater guys say is directly responsible for the double-dip recession suffered by many European countries). These ideas are not, of course, new either ("spend when you're losing money" is classic Keynesianism), but there is a sense that at least they are based on actual data and not The Model.
Reply With Quote
  #3  
Old 31 May 2014, 06:04 PM
Steve Steve is offline
 
Join Date: 19 October 2002
Location: Charleston, SC
Posts: 4,847
Default

Quote:
Originally Posted by Johnny Slick View Post
Sure, we're bad at seeing recessions in advance. I'd say that's at least as much due to bad zombie economics ideas being in vogue over the pat 30 years as anything else, but there is a deeper, worse idea at the core: a lot of freshwater economists argue that there is no such thing as recessions. That isn't to say that economies never contract 2 quarters in a row, it's that when this happens it's as close to random as anything can be, because if it wasn't random we'd have been better able to predict them by now.
Can you find someone who's argued that? Because off the top of my head I can't think of anyone who has and the idea that there are lots of economists in Chicago or Rochester saying that I'd be surprised.
Reply With Quote
  #4  
Old 31 May 2014, 07:38 PM
jimmy101_again jimmy101_again is offline
 
Join Date: 29 December 2005
Location: Greenwood, IN
Posts: 6,904
Default

Quote:
Originally Posted by Johnny Slick View Post
That isn't to say that economies never contract 2 quarters in a row, it's that when this happens it's as close to random as anything can be, because if it wasn't random we'd have been better able to predict them by now.
I think there is a logical fallacy or two in there somewhere.

If you can predict an event, and that event is bad, and there is anything you can do to avoid that event, then the act of predicting tends to be a self-cancelling prophecy. The prophecy is always wrong since people move to correct things before the event happens.

Secondly, the act of predicting (and publicizing or acting on that prediction) changes the system in a way that cannot be modeled very well. The classic example of this in financing are stock market prediction programs. If someone develops a highly accurate model that correctly predicts past behaviors (this has been done many times) and then applies that model to the current market then the current market is no longer the same as it was when the model was developed. Predicting and then acting on that prediction invalidates the model. You have to now develop a new model that takes all the previous factors into account plus the existence of the prediction system. Of course that second generation model also affects the system and is now also incorrect. Repeat until you can no longer convince anyone that modeling is a good idea.
Reply With Quote
  #5  
Old 31 May 2014, 11:19 PM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

Quote:
Originally Posted by Steve View Post
Can you find someone who's argued that? Because off the top of my head I can't think of anyone who has and the idea that there are lots of economists in Chicago or Rochester saying that I'd be surprised.
It's a consequence of Say's Law.

http://en.wikipedia.org/wiki/Say's_law
Reply With Quote
  #6  
Old 31 May 2014, 11:36 PM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

Quote:
Originally Posted by jimmy101_again View Post
I think there is a logical fallacy or two in there somewhere.

If you can predict an event, and that event is bad, and there is anything you can do to avoid that event, then the act of predicting tends to be a self-cancelling prophecy. The prophecy is always wrong since people move to correct things before the event happens.
1. Unless those parties in the position to be able to predict the bad event will benefit from it. For instance, the head of AIG had to have known that his act of taking out all those little bets guaranteeing the safety of so many packages of loans would, when those packages inevitably went south, not only bankrupt his company (at the time of their death, AIG had a bet-to-asset ratio north of 20:1, meaning that if only 5% of the crappy loan packages they were guaranteeing failed, they would be wiped out) but, because his company was guanteeing such a large portion of the market, the global economy would be savagely rent as well. He didn't care. He got his cash and still has nearly every penny of it.

That's kind of the point of how capitalism works: everybody competes to try to eke out the most money they can. Unfortunately, the system broke down the past time because much of the information which would have, for instance, stopped the housing boom in its tracks was obscured from investors.

I feel like I've gone off enough on this tangent but the point is this: the theory behind there being "no such thing as recessions" is basically that in an information-rich environment, any time the market collectively sees what looks like an economic slowdown, it will almost automatically adjust to it on its own long before the actual slowdown even occurs. In the case of the last recession, saying that it couldn't have existed is an argument based on a faulty premise; the obscured loan tranches, the dark traders, the bogus ratings given by ratings agencies apparently so far in the bag of Wall Street that they don't even know what corruption is anymore, and so on combined to restrict the flow of information to the market in unprecedented and unforseen ways (and, arguably, ways that would have been undoable if not for the deregulation of the markets under Reagan and Clinton).

2. All that being said about the *last* recession, I am not even making the claim that recessions are predictable. Right now, for whatever reason, they are not. Whether that's because of politics or a lack of knowledge or because they are somehow inherently unknowable in advance is an interesting question but besides the point of this. I am saying that they exist.

Quote:
Secondly, the act of predicting (and publicizing or acting on that prediction) changes the system in a way that cannot be modeled very well. The classic example of this in financing are stock market prediction programs. If someone develops a highly accurate model that correctly predicts past behaviors (this has been done many times) and then applies that model to the current market then the current market is no longer the same as it was when the model was developed. Predicting and then acting on that prediction invalidates the model. You have to now develop a new model that takes all the previous factors into account plus the existence of the prediction system. Of course that second generation model also affects the system and is now also incorrect. Repeat until you can no longer convince anyone that modeling is a good idea.
This would be another reason why recessions might exist even though economists can't predict them, sure.
Reply With Quote
  #7  
Old 01 June 2014, 12:15 AM
ganzfeld's Avatar
ganzfeld ganzfeld is offline
 
Join Date: 05 September 2005
Location: Kyoto, Japan
Posts: 23,577
Driver

It's a bit like having a driver predict his own car crash. The reason it happens is partly because it's unpredictable.
Reply With Quote
  #8  
Old 01 June 2014, 01:09 AM
geminilee's Avatar
geminilee geminilee is offline
 
Join Date: 02 December 2005
Location: New Orleans, LA
Posts: 11,518
Default

The weather analogy is a good one. Economics is a complex system (even if it does turn out to follow simple rules when broken down instance by instance) and thus chaos theory applies. You can't predict complex systems more than a few steps out, because there are too many variables.

This does not mean that economics* is a worthless field, but it does mean that predictions become less and less reliable the further out they are.


* Or meteorology, or evolution, or any other study of complex systems.
Reply With Quote
  #9  
Old 01 June 2014, 04:46 AM
ganzfeld's Avatar
ganzfeld ganzfeld is offline
 
Join Date: 05 September 2005
Location: Kyoto, Japan
Posts: 23,577
Default

Well, it's sort of like weather in which the behavior of not only thousands of weather forecasters (ie. prospective economists) but that of hundreds of thousands of similar individuals (some human and, these days, many not human) influences that weather greatly. It's a world in which a hurricane can occur in the blink of an eye simply because it was imagined or by some other quirk of a complex interacting weather systems of hundreds of thousands of different planets.
Reply With Quote
  #10  
Old 01 June 2014, 05:11 AM
geminilee's Avatar
geminilee geminilee is offline
 
Join Date: 02 December 2005
Location: New Orleans, LA
Posts: 11,518
Default

Even in economics, action has to follow imagination for it to have an effect, and it is that action which results in the changes.

My main point is that it is a system where many factors, all interacting with each other, result in any given outcome. Those defy prediction, at least greater than a few steps out.
Reply With Quote
  #11  
Old 01 June 2014, 06:11 AM
ganzfeld's Avatar
ganzfeld ganzfeld is offline
 
Join Date: 05 September 2005
Location: Kyoto, Japan
Posts: 23,577
Default

I just mean to point out that it's also predicting human (and non-human intelligent systems because they run a lot of stuff these days) behavior. Only it's about predicting the behavior of millions. So I think it's orders of magnitude more complex than predicting the weather - maybe even more complex than predicting things like earthquakes.
Reply With Quote
  #12  
Old 01 June 2014, 06:33 AM
geminilee's Avatar
geminilee geminilee is offline
 
Join Date: 02 December 2005
Location: New Orleans, LA
Posts: 11,518
Default

Well, yes. There are a great many more variables for one thing, and human behavior rarely boils down to simple laws in the way physical processes do. No analogy is perfect. The complexity/chaos theory issue is the main point.
Reply With Quote
  #13  
Old 01 June 2014, 07:55 AM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

Quote:
Originally Posted by ganzfeld View Post
I just mean to point out that it's also predicting human (and non-human intelligent systems because they run a lot of stuff these days) behavior. Only it's about predicting the behavior of millions. So I think it's orders of magnitude more complex than predicting the weather - maybe even more complex than predicting things like earthquakes.
I think you're grossly underestimating how scientists study weather.
Reply With Quote
  #14  
Old 01 June 2014, 08:42 AM
ganzfeld's Avatar
ganzfeld ganzfeld is offline
 
Join Date: 05 September 2005
Location: Kyoto, Japan
Posts: 23,577
Default

It doesn't matter how complex re the models they use to study it because the economy changes every time it's studied by every person involved in it, which the weather doesn't. Furthermore, the economy includes the weather (among many other extremely complex systems). In any case, I'm talking about the complexity of the problem, not the complexity of the models.
Reply With Quote
  #15  
Old 01 June 2014, 03:52 PM
jimmy101_again jimmy101_again is offline
 
Join Date: 29 December 2005
Location: Greenwood, IN
Posts: 6,904
Default

The problem with the weather analogy is that no matter what or how you predict the weather what is going to happen is going to happen. You can't do anything about a predicted hurricane that will change the chances of the hurricane occurring. If you predict or don't predict the chances of it happening are exactly the same.

In financial forecasting the prediction changes the outcome.
Reply With Quote
  #16  
Old 03 June 2014, 11:19 AM
Steve Steve is offline
 
Join Date: 19 October 2002
Location: Charleston, SC
Posts: 4,847
Default

Quote:
Originally Posted by Johnny Slick View Post
It's a consequence of Say's Law.

http://en.wikipedia.org/wiki/Say's_law
Not really. Even your site says :
Quote:
Say's law was generally accepted throughout the 19th century, though modified to incorporate the idea of a "boom and bust" cycle.
But again, I'm puzzled who these economists are arguing that there aren't recessions. ETA: Are you talking about something like rational expectations or real business cycle theory?
Reply With Quote
  #17  
Old 04 June 2014, 06:13 PM
Dr. Winston O'Boogie's Avatar
Dr. Winston O'Boogie Dr. Winston O'Boogie is offline
 
Join Date: 23 February 2000
Location: Fox Lake, IL
Posts: 5,207
Default

Quote:
Originally Posted by geminilee View Post
The weather analogy is a good one.
The weather analogy is a terrible one. If we can accurately predict a hurricane, we can avoid some of the consequences - death and destruction - but we can not stop the hurricane. If we can accurately predict a recession, then we can modify the factors that would create the recession and stop the recession.
Reply With Quote
  #18  
Old 04 June 2014, 09:04 PM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

I'm not entirely sure that's the case either, since we can't predict recessions very well at all. I think what we can do is predict situations that recessions occur in, determine if we're in one of those, and then react accordingly, but as this is a human thing, there's no telling that if those actions are taken and a recession doesn't occur that it was actually halted, and to make matters worse, people who insist that the actions had nothing to do with recessions will continue to say that.

A great example of this over the past 5 years has been how the US has managed to (narrowly) avoid double-dipping the way that much of Europe has. A lot of non-freshwater economists say this is because, unlike Europe, the US chose not to drastically cut spending to prioritize balancing a budget over economic growth. In fact, not only did we not cut spending, we put together a one-time spending package (which some saltwater guys - Paul Krugman in particular - say wasn't nearly large enough, but that's another story) to help spur the economy on and on top of that have pushed money into the market in an effort to get banks to spend it. People on the other side insist that either the recession is still coming (I know I've seen a lot of freak-out posts over QE, for instance, causing sudden hyper-inflation... some day) or else it wouldn't have happened anyway (and those especially firm believers in The Model say "well, yeah, but that's because there was no recession in the first place").

That situation aside, I think that overall I'm not sure that we're at a point where we can answer the question of how do we prevent recessions, or if when we do finally figure out what we can and can't predict in economics, if recessions will not be on that list. To a certain point I think that asking economists to predict and then prevent recessions may be as fantastical as asking those mapping the human genome to cure and/or prevent cancer.
Reply With Quote
  #19  
Old 04 June 2014, 09:07 PM
Johnny Slick's Avatar
Johnny Slick Johnny Slick is offline
 
Join Date: 13 February 2003
Location: Phoenix, AZ
Posts: 11,628
Default

Quote:
Originally Posted by Steve View Post
Not really. Even your site says :But again, I'm puzzled who these economists are arguing that there aren't recessions. ETA: Are you talking about something like rational expectations or real business cycle theory?
I'm not an economist but I believe they come from the rational expectations idea, that essentially there can't be recessions because the market would see that coming and correct for it in advance. Any economic downturns or ramp-ups are the result of something beyond the scope of economics, at least according to the model.

I wish I could provide names... I know there was a (decidedly freshwater) economist on Planet Money with this opinion last fall (he might even have been a U of Chicago guy, I don't remember).
Reply With Quote
  #20  
Old 05 June 2014, 02:00 AM
Steve Steve is offline
 
Join Date: 19 October 2002
Location: Charleston, SC
Posts: 4,847
Default

Quote:
Originally Posted by Dr. Winston O'Boogie View Post
The weather analogy is a terrible one. If we can accurately predict a hurricane, we can avoid some of the consequences - death and destruction - but we can not stop the hurricane. If we can accurately predict a recession, then we can modify the factors that would create the recession and stop the recession.
But the analogy was meant to show why recessions, like hurricanes, are difficult to predict. It had nothing to do with the consequences of making such predictions.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
The Astonishing Decline of Homelessness in America Simply Madeline Soapbox Derby 22 28 August 2013 08:08 PM
Paleontologists have discovered the oldest complete primate skeleton A Turtle Named Mack Techno-Babble 3 06 June 2013 01:41 AM
Panadol causes kidney failure snopes Inboxer Rebellion 7 05 November 2007 08:26 AM
Astonishing, any way you slice it snopes Food 19 06 August 2007 06:03 PM
Boeing 777 Engine Failure at Takeoff? JoeBentley Fauxtography 20 29 January 2007 03:55 PM


All times are GMT. The time now is 12:40 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2018, vBulletin Solutions, Inc.